Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    formulating unpredictable hypotheses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    i
    1.49
    ၤ
    1.24
     meninggal
    1.23
    ه
    1.19
    ।
    1.15
    x
    1.13
    constexpr
    1.12
    Т
    1.12
    ᗰ
    1.12
    ⚪
    1.10
    POSITIVE LOGITS
    theless
    1.59
    ्स
    1.55
    1.55
    てください
    1.50
    ly
    1.49
    ing
    1.45
    った
    1.40
    า
    1.36
    ors
    1.29
    ことで
    1.28
    Act Density 0.334%

    No Known Activations